The vertical-pod-autoscaler module: custom resources

Deprecated resource. Support for the resource might be removed in a later release.

Configuration for a vertical pod autoscaler, which automatically manages pod resources based on historical and real time resource utilization.

• apiVersion

  string

  Defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values.

  More information in the documentation.

• kind

  string

  Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase.

  More information in the documentation.

• metadata

  object
• spec

  object

  Required value

  Specification of the behavior of the autoscaler.

  More information in the documentation.

  • spec.recommenders

    array of objects

    Recommender responsible for generating recommendation for this object. List should be empty (then the default recommender will generate the recommendation) or contain exactly one recommender.

    VerticalPodAutoscalerRecommenderSelector points to a specific Vertical Pod Autoscaler recommender. In the future it might pass parameters to the recommender.

    • spec.recommenders.name

      string

      Required value

      Name of the recommender responsible for generating recommendation for this object.

  • spec.resourcePolicy

    object

    Controls how the autoscaler computes recommended resources. The resource policy may be used to set constraints on the recommendations for individual containers.

    If any individual containers need to be excluded from getting the VPA recommendations, then it must be disabled explicitly by setting mode to Off under containerPolicies.

    If not specified, the autoscaler computes recommended resources for all containers in the pod, without additional constraints.

    • spec.resourcePolicy.containerPolicies

      array of objects

      Per-container resource policies.

      Controls how autoscaler computes the recommended resources for a specific container.

      • spec.resourcePolicy.containerPolicies.containerName

        string

        Name of the container or DefaultContainerResourcePolicy, in which case the policy is used by the containers that don’t have their own policy specified.

      • spec.resourcePolicy.containerPolicies.controlledResources

        array of strings

        Specifies the type of recommendations that will be computed (and possibly applied) by VPA. If not specified, the default of (ResourceCPU, ResourceMemory) will be used.

        Default: ["cpu","memory"]

        • Element of the array

          string

          ResourceName is the name identifying various resources in a ResourceList.

      • spec.resourcePolicy.containerPolicies.controlledValues

        string

        Specifies which resource values should be controlled.

        Default: RequestsAndLimits

        Allowed values: RequestsAndLimits, RequestsOnly

      • spec.resourcePolicy.containerPolicies.maxAllowed

        object

        Specifies the maximum amount of resources that will be recommended for the container. The default is no maximum.

      • spec.resourcePolicy.containerPolicies.minAllowed

        object

        Specifies the minimal amount of resources that will be recommended for the container. The default is no minimum.

      • spec.resourcePolicy.containerPolicies.mode

        string

        Whether autoscaler is enabled for the container.

        Default: Auto

        Allowed values: Auto, Off

  • spec.scope

    string

    Scope controls additional recommendation partitioning for DaemonSet targets.

    The value must be a node label key (for example: node.kubernetes.io/instance-type).

  • spec.targetRef

    object

    Required value

    Points to the controller managing the set of pods for the autoscaler to control — e.g. Deployment, StatefulSet.

    VerticalPodAutoscaler can be targeted at controller implementing scale subresource (the pod set is retrieved from the controller’s ScaleStatus) or some well known controllers (e.g. for DaemonSet the pod set is read from the controller’s spec).

    If VerticalPodAutoscaler cannot use specified target it will report ConfigUnsupported condition. Note that VerticalPodAutoscaler does not require full implementation of scale subresource — it will not use it to modify the replica count. The only thing retrieved is a label selector matching pods grouped by the target resource.

    • spec.targetRef.apiVersion

      string

      API version of the referent.

    • spec.targetRef.kind

      string

      Required value

      Kind of the referent.

      More info.

    • spec.targetRef.name

      string

      Required value

      Name of the referent.

      More info.

  • spec.updatePolicy

    object

    Describes the rules on how changes are applied to the pods. If not specified, all fields in the PodUpdatePolicy are set to their default values.

    • spec.updatePolicy.evictionRequirements

      array of objects

      EvictionRequirements is a list of EvictionRequirements that need to evaluate to true in order for a Pod to be evicted. If more than one EvictionRequirement is specified, all of them need to be fulfilled to allow eviction.

      EvictionRequirement defines a single condition which needs to be true in order to evict a Pod.

      • spec.updatePolicy.evictionRequirements.changeRequirement

        string

        Required value

        EvictionChangeRequirement refers to the relationship between the new target recommendation for a Pod and its current requests, what kind of change is necessary for the Pod to be evicted.

        Allowed values: TargetHigherThanRequests, TargetLowerThanRequests

      • spec.updatePolicy.evictionRequirements.resources

        array of strings

        Required value

        Resources is a list of one or more resources that the condition applies to. If more than one resource is given, the EvictionRequirement is fulfilled if at least one resource meets changeRequirement.

        • Element of the array

          string

          ResourceName is the name identifying various resources in a ResourceList.

    • spec.updatePolicy.minReplicas

      integer

      Minimal number of replicas which need to be alive for Updater to attempt pod eviction (pending other checks like PDB). Only positive values are allowed. Overrides global ‘–min-replicas’ flag.

    • spec.updatePolicy.updateMode

      string

      The operation mode of the VPA controller.

      • Auto — currently, Auto and Recreate modes do the same thing. This mode is reserved for the Pod inplace resource update in Kubernetes.
      • Recreate — this mode allows VPA to modify resource requirements of the running pods (i.e., restart them during operation). This mode might result in temporary unavailability of the service (due to restart) if you have just one replica (replicas: 1). In this mode, VPA does not recreate pods that were not created by a controller.
      • Initial — VPA modifies pod resources only when pods are started (but not when pods are running).
      • Off — VPA does not take any action to update the resource requests for the running containers. Autoscaler calculates recomendations and stores them in the status field of the VPA object. You can browse VPA recommendations using the following command: d8 k describe vpa <vpa-name>.
      • InPlaceOrRecreate (GA) — supported in Kubernetes 1.33 and later. The VPA attempts to update pod resources without a restart; if that is not possible, it recreates the pod.

      Default: InPlaceOrRecreate

      Allowed values: Off, Initial, Recreate, Auto, InPlaceOrRecreate

• status

  object

  Current information about the autoscaler.

  • status.conditions

    array of objects

    Set of conditions required for this autoscaler to scale its target, and indicates whether or not those conditions are met.

    Describes the state of a VerticalPodAutoscaler at a certain point.

    • status.conditions.lastTransitionTime

      string

      Last time the condition transitioned from one status to another.

    • status.conditions.message

      string

      A human-readable explanation containing details about the transition.

    • status.conditions.reason

      string

      Reason for the condition’s last transition.

    • status.conditions.status

      string

      Required value

      Status of the condition (True, False, Unknown).

    • status.conditions.type

      string

      Required value

      Describes the current condition.

  • status.groups

    array of objects

    Contains grouped recommendations by scope value when spec.scope is used.

    • status.groups.containerRecommendations

      array of objects

      Resources recommended by autoscaler for each container in this group.

      • status.groups.containerRecommendations.containerName

        string

        Name of the container.

      • status.groups.containerRecommendations.lowerBound

        object

        Minimum recommended amount of resources. Observes ContainerResourcePolicy. This amount is not guaranteed to be sufficient for the application to operate in a stable way, however running with less resources is likely to have significant impact on performance/availability.

      • status.groups.containerRecommendations.scope

        string

        Scope binds this recommendation to a specific scope value in form “=".

      • status.groups.containerRecommendations.target

        object

        Required value

        Recommended amount of resources. Observes ContainerResourcePolicy.

      • status.groups.containerRecommendations.uncappedTarget

        object

        The most recent recommended resources target computed by the autoscaler for the controlled pods, based only on actual resource usage, not taking into account the ContainerResourcePolicy.

        May differ from the Recommendation if the actual resource usage causes the target to violate the ContainerResourcePolicy (lower than MinAllowed or higher that MaxAllowed). Used only as status indication, will not affect actual resource assignment.

      • status.groups.containerRecommendations.upperBound

        object

        Maximum recommended amount of resources.

        Observes ContainerResourcePolicy. Any resources allocated beyond this value are likely wasted. This value may be larger than the maximum amount of application is actually capable of consuming.

    • status.groups.scopeValue

      string

      Required value

      Scope value of spec.scope label key for this recommendation group.

  • status.recommendation

    object

    The most recently computed amount of resources recommended by the autoscaler for the controlled pods.

    • status.recommendation.containerRecommendations

      array of objects

      Resources recommended by the autoscaler for each container.

      Recommendation of resources computed by autoscaler for a specific container. Respects the container resource policy if present in the spec. In particular the recommendation is not produced for containers with ContainerScalingMode set to ‘Off’.

      • status.recommendation.containerRecommendations.containerName

        string

        Name of the container.

      • status.recommendation.containerRecommendations.lowerBound

        object

        Minimum recommended amount of resources. Observes ContainerResourcePolicy. This amount is not guaranteed to be sufficient for the application to operate in a stable way, however running with less resources is likely to have significant impact on performance/availability.

      • status.recommendation.containerRecommendations.scope

        string

        Scope binds this recommendation to a specific scope value in form “=".

      • status.recommendation.containerRecommendations.target

        object

        Required value

        Recommended amount of resources. Observes ContainerResourcePolicy.

      • status.recommendation.containerRecommendations.uncappedTarget

        object

        The most recent recommended resources target computed by the autoscaler for the controlled pods, based only on actual resource usage, not taking into account the ContainerResourcePolicy. May differ from the Recommendation if the actual resource usage causes the target to violate the ContainerResourcePolicy (lower than MinAllowed or higher that MaxAllowed). Used only as status indication, will not affect actual resource assignment.

      • status.recommendation.containerRecommendations.upperBound

        object

        Maximum recommended amount of resources. Observes ContainerResourcePolicy. Any resources allocated beyond this value are likely wasted. This value may be larger than the maximum amount of application is actually capable of consuming.